C1W3 Assignment: Callback Function

Course Q&A TensorFlow: Data and Deployment Specialization Browser-based Models with TensorFlow.js
1

The if clause on the callback function looks like this:

if(logs.get(‘acc’)>0.7 and logs.get(‘val_acc’)>0.7):
print(“\nReached 99.8% accuracy so cancelling training!”)
self.model.stop_training = True

I don’t know which part is wrong, either the if condition that both accuracies have to be only above 70% or the printout statement, saying we reached 99.8%???

In my case, the training stopped already after 3 epochs.

Did anyone have had an issue with that before?

So the model passed the grader, so at least this is not an issue

2

Peter,
have you chosen the right course and specialisation. if not please correct it. So respective mentors can respond.

Otherwise. Share this part of the code via DM

3

Yes I think I am in the correct course.

The callback function is in the C1_W3_Assignment.ipynb:
Converting a Keras Model to JSON Format

and it is coded for us.

4

Okay my bad. I am sorry, this assignment requires you to convert the model to JSON format and that is the main assignment. Even for me it had 99% accuracy.

So that should not be a problem.